The intended joke is that hypervalent iodine compounds like Dess-Martin periodinane flip between different oxidation states like you often see for transition metals. As an example, the mechanism usually drawn for oxidations by DMP is similar to those drawn for, e.g., PCC/Jones reagent, where the electrons removed from the substrate is "banked" at the metal center. Obviously, redox chemistry is not at all limited to transition metals, but I am often surprised at iodine's propensity to engage in it. A lot of research over the past decade or two has also developed redox catalysis with these reagents, reactivity which is commonly (though again not always) the purview of transition metals.
Physicists are notorious for approximating, and astronomers are even worse. But there are some subfields where they care about being more precise, and you maybe break the periodic table into a handful of elements plus alphas. And there's that one or two people getting exquisite spectral resolution and signal-to-noise on a few stars and measuring the abundance of Technetium or whatever.
It's why I fucking love astrophysics. There's so much handwaving because so much information is observed.
But without the handwaving you can't find crazy ass things like nuclear fusion being behind the power of stars. You find these really big numbers everywhere that make the "normal stuff" negligible.
It not that the precision isn't important, it's just not always relevant at particular scales, like the scale of space.